Stéphane Karasiewicz — Senior Data Scientist with 7+ years delivering ML solutions from research to production. Specialized in environmental analytics, time-series modeling, and scalable pipelines.
I don't just build models — I deliver business outcomes. Every project starts with your goals and ends with measurable impact.
Custom ML models: churn prediction, price forecasting, anomaly detection. Full lifecycle from preprocessing to production deployment.
Satellite data analysis, climate risk modeling, ESG reporting, and environmental monitoring. 2TB+ satellite data management experience.
Interactive analytics dashboards with country-level maps, KPI tracking, and real-time monitoring. 5+ production dashboards delivered.
End-to-end ML infrastructure: automated training, CI/CD, Docker containerization, cloud deployment, and ETL pipelines at scale.
Text classification, document extraction, image recognition using state-of-the-art transformer and CNN architectures.
Advanced multivariate analysis, time-series modeling, and custom R package development. Published author on CRAN.
A track record spanning marine research, freelance consulting, and industry ML engineering.
Raincoat LLC — Peril Risk Assessment
IFREMER, Boulogne-sur-Mer
IFREMER Brest & Boulogne — Satellite & Marine MLReal deliverables, real impact — not just notebooks.
Production-grade ML pipeline for credit card fraud detection with real-time streaming inference, SHAP explainability, A/B testing, and a Streamlit monitoring dashboard.
→ Multi-page real-time dashboardReal-time sentiment monitoring using fine-tuned RoBERTa, BERTopic for topic modeling, entity-level sentiment with spaCy NER, and a Plotly Dash dashboard.
→ Real-time multi-model NLP pipelineU-Net architecture in TensorFlow for medical image segmentation with data augmentation, MC Dropout uncertainty, DICOM handling, and Grad-CAM visualization.
→ ONNX-exported production modelDeep learning system for land use classification and change detection from satellite imagery using PyTorch with ResNet/EfficientNet and interactive Folium maps.
→ Multi-temporal change detectionInterpretable ML model for loan approval using LightGBM with SHAP/LIME explainability, Fairlearn bias testing, traditional WoE/IV scorecard, and regulatory compliance.
→ Fairness-audited production modelRetrieval-augmented generation chatbot ingesting PDF/DOCX/Markdown, embedding into ChromaDB, answering with citations using LangChain and Claude/OpenAI APIs.
→ Multi-format document Q&A systemContributing to science and the data community through packages and publications.
We discuss your data, goals and constraints. Free, 15 min, no commitment.
Clear approach, timeline, deliverables, and pricing — fixed or time-and-materials.
Weekly check-ins, shared repos, transparent progress. You're never in the dark.
Polished deliverables, documentation, and handoff support to your team.
Book a free 15-minute discovery call. No pitch, no pressure — just an honest conversation about how data science can help your business.
Book a Discovery Call → Email Me